Calculation Algorithm of Upward Longwave Radiation Based on Surface Types

Upward longwave radiation (ULR) is an important element for climate change analysis. We calculated ULR using land surface temperature (LST), sea surface temperature (SST), and downward longwave radiation (DLR) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Cloud and the Earth's Radiant Energy System (CERES), along with broadband emissivity calculated using four multiple linear regression models designed to consider specific land-cover categories along with MODIS data and reflectivity data for 241 materials from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). Our broadband emissivity values were compared with those of Wang et al. Journal of Geophysical Research: Atmospheres, 110(D11), (2005); all correlation coefficients were higher than 0.90 for the period from January to December 2016, and the root mean square error (RMSE) was 0.00750-0.00825. We then calculated ULR using the broadband emissivity and compared the results with CERES data, resulting in smaller errors relative to CERES than found by Wang et al. (2005). Furthermore, the calculated ULR compared with the observational data of Baseline Surface Radiation Network (BSRN) and the resultant statistical values of R(correlation coefficient), RMSE, and bias were 0.93, 21.22 W m(-2), and 0.29 respectively.